We have previously identified TGFBR1*6A, a common variant of the TGFBR1 gene, and shown that it transmits TGF-a growth inhibitory signals less effectively than TGFBR1. Our recent meta-analyses show that TGFBR1*6A carriers have a significantly increased risk of colon, breast and ovarian cancer as compared with non-carriers. Overall, cancer risk is increased by 19% among heterozygotes and 70% among homozygotes, a pattern indicative of an allelic dosing effect. We have also shown that TGFBR1*6A may contribute to hereditary colorectal cancer. More than one in eight healthy individuals and one in six patients with cancer is a TGFBR1*6A carrier, which establishes TGFBR1*6A as the first high-frequency low-penetrance candidate tumor susceptibility allele. In contrast, increased TGF-a circulating levels have been associated with a decreased cancer risk in animal models. A common Leucine to Proline substitution at the 10th amino acid position variant within the human TGF- a1 (TGFB1) gene results in higher in vitro extracellular TGFB1 secretion. Carriers of the TGFB1*CC genotype have higher in vivo TGFB1 circulating levels than carriers of the TGFB1*TT genotype. TGFBR1 and TGFB1 variants my have opposite or synergistic effects on colorectal cancer risk. Our central hypothesis is that a combined assessment of the two functionally-relevant TGF- a pathway signaling variants will predict colorectal cancer risk more accurately than each variant alone. The NCI-sponsored familial colorectal cancer registry is an ideal resource in which to test this hypothesis. Using a sibling-matched case-control design we will genotype a total of 4,208 full sibling case-control pairs and First: assess the association between TGFBR1*6A and colorectal cancer. Second: assess the association between TGFB1 and colorectal cancer and perform haplotype analysis of the TGFB1 gene; Third: analyze gene-gene interactions between TGFBR1 and TGFB1. This will explore the relationships between the two functional TGF-a pathway polymorphisms and colorectal risk and determine whether TGF- a signaling, as predicted by these two variants, is associated with colorectal cancer risk; and, Fourth: investigate the relationship between TGF-a pathway polymorphisms and tumor microsatellite instability. [unreadable] [unreadable] [unreadable]